Celecoxib to improve scar quality following acute burn injury: Lessons learned after premature termination of a randomised trial

•The use of non-steroidal anti-inflammatory drugs (NSAIDs) following burn injury is supported by pre-clinical animal studies.•Physiological effects of NSAIDs have been demonstrated in burn injured humans but the effect on wound and scar outcomes is unknown.•Participants allocated to celecoxib had no significant improvement in scar quality compared to those receiving placebo, although early study termination limits confidence in this finding.•Flexible follow up and digital health innovation offer solutions to some of the challenges encountered when studying outcomes of non-major burn injury. Pre-clinical studies suggest that non-steroidal anti-inflammatory therapy may reduce acute inflammation and modulate scar formation after burn injury. The objective of this trial was to determine if celecoxib administered soon after acute burn improved scar quality. A single centre, parallel group, placebo-controlled, randomised, superiority trial. Adults with non-major acute burn injury were eligible for recruitment. Participants within 48 h of admission for acute burn injury received either six weeks of celecoxib 200 mg twice daily or identically packaged placebo capsules. The primary outcome was participant-reported Patient Observer Scar Assessment Scale (POSAS) at day 42 following burn injury. The best possible score using this assessment is 7 and the worst 70. Key safety outcomes included study drug side effects. Restrictions imposed during the COVID-19 pandemic resulted in slow recruitment and early termination after 59 of the 150 planned participants were enrolled. The primary outcome was available for 21/30 (70 %) participants in the celecoxib arm and for 23/29 (79 %) who received placebo. There was no significant difference in the primary outcome between treatment groups with a mean POSAS of 33.4 (standard deviation 12.7) in the celecoxib arm and 36.9 (13.7) in the control arm giving a difference of −3.49 (95 % CI [-11.57, 4.59], p =.39). Gastrointestinal symptoms occurred in 3/30 (10 %) allocated celecoxib and in none allocated placebo. Secondary outcomes including graft loss and delayed healing were similar between groups. Anti-inflammatory therapy did not improve scar quality. Due to early trial termination and significant loss to follow up these findings should be interpreted cautiously. To improve generalisability and to attain recruitment targets, future trials should apply different approaches to improve participant retention as well as including patient